hobbes VS sexp

https://github.com/morganstanley/hobbes

Things like structural equality and ordering are user functions here rather than magic internal definitions (e.g. equality is defined as a type class, and type class instances can deconstruct algebraic types at compile time to decide how to implement instances). But evaluation is eager by default, so it's pretty easy to reason about performance. And data structures can be persisted to files and/or shared transparently in memory with C/C++ programs without translation, also very convenient for the places where this was used. Actually we built some very big and complicated time series databases with it, used in both pre and post trade settings where ~15% of daily US equity trades happen. So I think these observations are useful and have passed through some pretty significant real tests.

Have a look at Hobbes from Morgan Stanley

https://github.com/morganstanley/hobbes

This approach has several benefits, but there's a lot of appeal also to the way they're doing it, to have the same PL across stages (and then maybe have more than just two stages).

There's a tool sexp that Jane Street uses internally to slice and dice sexp logs, but, yes, I get that there aren't logging services out there that support sexps. They should, though, as sexps have unique advantages over JSON. They tend to be more human-readable too. Great blog post overall!